Informacja

Drogi użytkowniku, aplikacja do prawidłowego działania wymaga obsługi JavaScript. Proszę włącz obsługę JavaScript w Twojej przeglądarce.

English (EN)·Polski (PL)·Yкраїнський (UK)
Przejdź do strony domowej biblioteki Centralna Biblioteka Wojskowa Link otwiera się w nowym oknie Logo biblioteki Prolib Integro - strona głównaCentralna Biblioteka Wojskowa

Wyszukujesz frazę "evolutionary learning" wg kryterium: Temat

  Lista filtrów
Wyrównaj
Wyświetlanie 1-10 z 20
Tytuł:
A biologically inspired approach to feasible gait learning for a hexapod robot
Autorzy:
Belter, D.
Skrzypczyński, P.
Powiązania:
https://bibliotekanauki.pl/articles/907777.pdf
Data publikacji:
2010
Wydawca:
Uniwersytet Zielonogórski. Oficyna Wydawnicza
Tematy:
identyfikacja modelu
uczenie się ewolucyjne
robot nożny
evolutionary learning
legged robots
gait generation
model identification
reality gap
Opis:
The objective of this paper is to develop feasible gait patterns that could be used to control a real hexapod walking robot. These gaits should enable the fastest movement that is possible with the given robot's mechanics and drives on a flat terrain. Biological inspirations are commonly used in the design of walking robots and their control algorithms. However, legged robots differ significantly from their biological counterparts. Hence we believe that gait patterns should be learned using the robot or its simulation model rather than copied from insect behaviour. However, as we have found tahula rasa learning ineffective in this case due to the large and complicated search space, we adopt a different strategy: in a series of simulations we show how a progressive reduction of the permissible search space for the leg movements leads to the evolution of effective gait patterns. This strategy enables the evolutionary algorithm to discover proper leg co-ordination rules for a hexapod robot, using only simple dependencies between the states of the legs and a simple fitness function. The dependencies used are inspired by typical insect behaviour, although we show that all the introduced rules emerge also naturally in the evolved gait patterns. Finally, the gaits evolved in simulations are shown to be effective in experiments on a real walking robot.
Źródło:
International Journal of Applied Mathematics and Computer Science; 2010, 20, 1; 69-84
1641-876X
2083-8492
Pojawia się w:
International Journal of Applied Mathematics and Computer Science
Dostawca treści:
Biblioteka Nauki
Artykuł
Tytuł:
Examining the impact of positive and negative constant learning on the evolution rate
Autorzy:
Gajer, M.
Powiązania:
https://bibliotekanauki.pl/articles/1943200.pdf
Data publikacji:
2009
Wydawca:
Politechnika Gdańska
Tematy:
evolutionary systems
learning process
constant learning
Opis:
The paper discusses the influence of learning on evolutionary processes. In biological sciences it is a well-known fact that the rate of evolution can be effected by learning and the same phenomena can also be observed in artificial evolutionary systems, however, their nature is still not sufficiently well understood. In the paper the influence of constant learning on the rate of evolution is examined. The constant learning is a kind of learning during which the genotype of the individual being taught is moved toward the global optimum over a constant value. If the fitness function is monotonic, it can be concluded from the mathematical theory that such kind of learning should decelerate evolution. However, this fact is highly counterintuitive and for this reason it should be proved by numerical experiments. In the article the results of numerical simulations are presented. They prove that evolution is indeed decelerated by learning in case of the sigmoid fitness function. Moreover, two cases of constant learning were examined in the paper. These are the positive and negative constant learning. It was demonstrated that in the case of the negative constant learning the evolution was decelerated to a larger extent than in the case of the positive constant learning. The obtained results can help explain certain phenomena concerning the impact of learning on the evolution both in natural and artificial evolutionary systems.
Źródło:
TASK Quarterly. Scientific Bulletin of Academic Computer Centre in Gdansk; 2009, 13, 4; 355-362
1428-6394
Pojawia się w:
TASK Quarterly. Scientific Bulletin of Academic Computer Centre in Gdansk
Dostawca treści:
Biblioteka Nauki
Artykuł
Tytuł:
Learning decision rules using a distributed evolutionary algorithm
Autorzy:
Kwedlo, W.
Krętowski, M.
Powiązania:
https://bibliotekanauki.pl/articles/1986918.pdf
Data publikacji:
2002
Wydawca:
Politechnika Gdańska
Tematy:
decision rule learning
distributed evolutionary algorithms
Opis:
A new parallel method for learning decision rules from databases by using an evolutionary algorithm is proposed. We describe an implementation of EDRL-MD system in the cluster of multiprocessor machines connected by Fast Ethernet. Our approach consists in a distribution of the learning set into processors of the cluster. The evolutionary algorithm uses a master-slave model to compute the fitness function in parallel. The remiander of evolutionary algorithm is executed in the master node. The experimental results show, that for large datasets our approach is able to obtain a significant speed-up in comparison to a single processor version.
Źródło:
TASK Quarterly. Scientific Bulletin of Academic Computer Centre in Gdansk; 2002, 6, 3; 483-492
1428-6394
Pojawia się w:
TASK Quarterly. Scientific Bulletin of Academic Computer Centre in Gdansk
Dostawca treści:
Biblioteka Nauki
Artykuł
Tytuł:
LCS Approach to Tasks Scheduling Problem in the Two Processor System
Autorzy:
Wasielewska, K.
Seredyński, F.
Powiązania:
https://bibliotekanauki.pl/articles/92950.pdf
Data publikacji:
2007
Wydawca:
Uniwersytet Przyrodniczo-Humanistyczny w Siedlcach
Tematy:
learning classifier systems
scheduling problem
evolutionary technique
Opis:
In this paper we propose an approach to solve multiprocessor scheduling problem with use of rule-based learning machine - Learning Classifier System (LCS). LCS combines reinforcement learning and evolutionary computing to produce adaptive systems. We interpret the multiprocessor scheduling problem as multi-step problem, where a feedback is given after some number steps. We show that LCS is able to solve scheduling tasks of a parallel program in the two processor system.
Źródło:
Studia Informatica : systems and information technology; 2007, 2(9); 29-39
1731-2264
Pojawia się w:
Studia Informatica : systems and information technology
Dostawca treści:
Biblioteka Nauki
Artykuł
Tytuł:
Learning board evaluation function for Othello by hybridizing coevolution with temporal difference learning
Autorzy:
Szubert, M.
Jaśkowski, W.
Krawiec, K.
Powiązania:
https://bibliotekanauki.pl/articles/206175.pdf
Data publikacji:
2011
Wydawca:
Polska Akademia Nauk. Instytut Badań Systemowych PAN
Tematy:
evolutionary computation
coevolutionary algorithms
reinforcement learning
memetic computing
game strategy learning
Opis:
Hybridization of global and local search techniques has already produced promising results in the fields of optimization and machine learning. It is commonly presumed that approaches employing this idea, like memetic algorithms combining evolutionary algorithms and local search, benefit from complementarity of constituent methods and maintain the right balance between exploration and exploitation of the search space. While such extensions of evolutionary algorithms have been intensively studied, hybrids of local search with coevolutionary algorithms have not received much attention. In this paper we attempt to fill this gap by presenting Coevolutionary Temporal Difference Learning (CTDL) that works by interlacing global search provided by competitive coevolution and local search by means of temporal difference learning. We verify CTDL by applying it to the board game of Othello, where it learns board evaluation functions represented by a linear architecture of weighted piece counter. The results of a computational experiment show CTDL superiority compared to coevolutionary algorithm and temporal difference learning alone, both in terms of performance of elaborated strategies and computational cost. To further exploit CTDL potential, we extend it by an archive that keeps track of selected well-performing solutions found so far and uses them to improve search convergence. The overall conclusion is that the fusion of various forms of coevolution with a gradient-based local search can be highly beneficial and deserves further study.
Źródło:
Control and Cybernetics; 2011, 40, 3; 805-831
0324-8569
Pojawia się w:
Control and Cybernetics
Dostawca treści:
Biblioteka Nauki
Artykuł
Tytuł:
Evolutionary algorithm for learning Bayesian structures from data
Autorzy:
Kozłowski, M.
Wierzchoń, S. T.
Powiązania:
https://bibliotekanauki.pl/articles/1986916.pdf
Data publikacji:
2002
Wydawca:
Politechnika Gdańska
Tematy:
Bayesian networks
structure learning
evolutionary algorithm
discrete optimization
Opis:
In this paper we report an evolutionary approach to learning Bayesian networks from data. We explain reasons, which advocate such a non-deterministic approach. We analyze weaknesses of previous works and come to conclusion that we should operate in the search space native for the problem i.e. in the space of directed acyclic graphs instead of standard space of binary strings. This requires adaptation of evolutionary methodology into very specific needs. We propose quite new data representation and implementation of generalized genetic operators and then we present an efficient algorithm capable of learning complex networks without additional assumptions. We discuss results obtained with this algorithm. The approach presented in this paper can be extended with the possibility to absorb some suggestions from experts or obtained by means of data preprocessing.
Źródło:
TASK Quarterly. Scientific Bulletin of Academic Computer Centre in Gdansk; 2002, 6, 3; 509-521
1428-6394
Pojawia się w:
TASK Quarterly. Scientific Bulletin of Academic Computer Centre in Gdansk
Dostawca treści:
Biblioteka Nauki
Artykuł
Tytuł:
A survey of big data classification strategies
Autorzy:
Banchhor, Chitrakant
Srinivasu, N.
Powiązania:
https://bibliotekanauki.pl/articles/2050171.pdf
Data publikacji:
2020
Wydawca:
Polska Akademia Nauk. Instytut Badań Systemowych PAN
Tematy:
big data
data mining
MapReduce
classification
machine learning
evolutionary intelligence
deep learning
Opis:
Big data plays nowadays a major role in finance, industry, medicine, and various other fields. In this survey, 50 research papers are reviewed regarding different big data classification techniques presented and/or used in the respective studies. The classification techniques are categorized into machine learning, evolutionary intelligence, fuzzy-based approaches, deep learning and so on. The research gaps and the challenges of the big data classification, faced by the existing techniques are also listed and described, which should help the researchers in enhancing the effectiveness of their future works. The research papers are analyzed for different techniques with respect to software tools, datasets used, publication year, classification techniques, and the performance metrics. It can be concluded from the here presented survey that the most frequently used big data classification methods are based on the machine learning techniques and the apparently most commonly used dataset for big data classification is the UCI repository dataset. The most frequently used performance metrics are accuracy and execution time.
Źródło:
Control and Cybernetics; 2020, 49, 4; 447-469
0324-8569
Pojawia się w:
Control and Cybernetics
Dostawca treści:
Biblioteka Nauki
Artykuł
Tytuł:
Decelerating the rate of evolution with constant learning
Spowalnianie tempa ewolucji z wykorzystaniem uczenia stałego
Autorzy:
Gajer, M.
Powiązania:
https://bibliotekanauki.pl/articles/275811.pdf
Data publikacji:
2012
Wydawca:
Sieć Badawcza Łukasiewicz - Przemysłowy Instytut Automatyki i Pomiarów
Tematy:
systemy ewolucyjne
proces uczenia się
uczenie stałe
efekt Baldwina
evolutionary systems
learning process
constant learning
Baldwin effect
Opis:
Evolution and learning are two main processes that are considered in the case of artificial intelligence and artificial life systems. These two processes can interact with each other, which is called the Baldwin effect. Especially, the introduction of learning process into an evolutionary system can cause acceleration or deceleration of the rate of evolution both in the case of artificial and natural evolutionary systems. However, there is still a lack of a solid mathematical theory that could thoroughly explain the phenomena concerned with the impact of learning on the rate of evolution. In the case of constant learning, that is a process during which individuals are moved a constant value toward the optimum, it was proved that if the second derivative of the logarithm of the fitness function is negative, the rate of the evolution should be slowed down as a result of the introduction of constant learning. In the paper we assume an evolutionary system with the asymptotic fitness function for which the theory states that the introduction of constant learning should lead to deceleration of the rate of evolution. The results of numerous computer simulations confirmed the theory and demonstrated that the deceleration of the rate of the evolution is significant. Moreover, the impact of the intensity of mutation on the degree of deceleration of the rate of evolution could also be observed.
Ewolucja i uczenie się są dwoma głównymi procesami rozpatrywanymi w kontekście systemów sztucznej inteligencji i systemów sztucznego życia. Oba wymienione procesy mogą wchodzić we wzajemną interakcję, co bywa określane mianem efektu Baldwina. W szczególności wprowadzenie procesu uczenia do systemu ewolucyjnego może powodować przyspieszenie bądź spowolnienie tempa ewolucji zarówno w przypadku sztucznych, jak i naturalnych systemów ewolucyjnych. Obecnie wciąż odczuwany jest brak solidnej teorii matematycznej, która byłaby w stanie wyjaśnić w pełni zjawiska związane z wpływem procesu uczenia na tempo przebiegu ewolucji. W przypadku tzw. uczenia stałego, które polega na systematycznym przesuwaniu o stałą wartość genotypu osobnika w kierunku poszukiwanego optimum, udowodniono, że jeżeli druga pochodna logarytmu funkcji dopasowania jest ujemna, wówczas tempo przebiegu ewolucji powinno ulec spowolnieniu w wyniku wprowadzenia do systemu ewolucyjnego uczenia stałego. W artykule rozważono system ewolucyjny z asymptotyczną funkcją dopasowania, w przypadku którego zgodnie z teorią wprowadzenie uczenia stałego powinno wywołać spowolnienie tempa przebiegu ewolucji. Liczne wyniki symulacji komputerowych potwierdzają przewidywania teorii i pokazują, że spowolnienie tempa ewolucji jest istotne. Ponadto można zaobserwować dodatkowy wpływ częstotliwości mutacji na spowolnienie tempa ewolucji.
Źródło:
Pomiary Automatyka Robotyka; 2012, 16, 11; 50-53
1427-9126
Pojawia się w:
Pomiary Automatyka Robotyka
Dostawca treści:
Biblioteka Nauki
Artykuł
Tytuł:
Analiza wpływu uczenia stałego na tempo przebiegu procesów ewolucyjnych
Analysis of the impact of constant learning on the evolution rate
Autorzy:
Gajer, M.
Powiązania:
https://bibliotekanauki.pl/articles/156926.pdf
Data publikacji:
2010
Wydawca:
Stowarzyszenie Inżynierów i Techników Mechaników Polskich
Tematy:
systemy ewolucyjne
proces uczenia
efekt Baldwina
evolutionary systems
learning process
Baldwin effect
Opis:
W artykule rozważono wpływ procesu uczenia na tempo zachodzenia przemian ewolucyjnych. Zjawisko polegające na tym, że wprowadzenie do sytemu ewolucyjnego procesu uczenia może zarówno przyspieszać, jak i spowalniać ewolucję, jest od dawna znane w naukach przyrodniczych i określane jest mianem efektu Baldwina. Natomiast brak jest ogólnej teorii opisującej rozważane zjawiska w sposób ilościowy. W artykule przedstawiono teoretyczną analizę wpływu uczenia stałego na tempo ewolucji. Uzyskane wyniki zostały dodatkowo potwierdzone przeprowadzonymi przez autora symulacjami numerycznymi, z których wynika, że w systemach ewolucyjnych z dodatnią i monotoniczną funkcją celu wprowadzenie uczenia stałego zawsze powoduje spowolnienie ewolucji.
The paper deals with the influence of learning on the evolution rate. It is a well-known fact that learning can under some circumstances accelerate or decelerate evolution, but there is no general theory that could explain these phenomena. The work [11] proposes a mathematical method with use of which one can determine whether the evolution will be accelerated or decelerated by learning for a monotonic and positive fitness function. This mathematical method is based on analysis of the fitness function logarithm second derivative. In the paper there is presented an experimental evolu-tionary system for which it was proved that the fitness function logarithm second derivative is negative. This fact causes that introduction of the constant learning to such a system must lead to deceleration of evolution. However, the mathematical method presented in [11] does not allow for any quantitative analysis of this phenomenon. Numerical experiments were conducted by the author of this paper in order to confirm the theoretical results obtained before. The simulation results of impact of learning on the evolution rate are shown in Figs. 1- 5. It can be noted that the deceleration of evolution, especially in the case of lower number of evolutionary algorithm generations, is relatively large. The impact of mutation intensity on the evolution rate was also examined. It was shown that increase in the mutation intensity accelerates the evolution significantly. The paper is organised as follows: Section 1 is the introduction, Section 2 presents the outline of the mathematical method based on gain function analysis, Section 3 discusses the results of numerical simulations, Section 4 gives the concluding remarks..
Źródło:
Pomiary Automatyka Kontrola; 2010, R. 56, nr 5, 5; 475-478
0032-4140
Pojawia się w:
Pomiary Automatyka Kontrola
Dostawca treści:
Biblioteka Nauki
Artykuł
Tytuł:
Machine Learning Methods in Algorithmic Trading Strategy Optimization – Design and Time Efficiency
Autorzy:
Ryś, Przemysław
Ślepaczuk, Robert
Powiązania:
https://bibliotekanauki.pl/articles/1356900.pdf
Data publikacji:
2019-08-09
Wydawca:
Uniwersytet Warszawski. Wydział Nauk Ekonomicznych
Tematy:
Algorithmic trading
investment strategy
machine learning
optimization
differential evolutionary method
cross-validation
overfitting
Opis:
The main aim of this paper was to formulate and analyse the machine learning methods, fitted to the strategy parameters optimization specificity. The most important problems are the sensitivity of a strategy performance to little parameter changes and numerous local extrema distributed over the solution space in an irregular way. The methods were designed for the purpose of significant shortening of the computation time, without a substantial loss of strategy quality. The efficiency of methods was compared for three different pairs of assets in case of moving averages crossover system. The problem was presented for three sets of two assets’ portfolios. In the first case, a strategy was trading on the SPX and DAX index futures; in the second, on the AAPL and MSFT stocks; and finally, in the third case, on the HGF and CBF commodities futures. The methods operated on the in-sample data, containing 16 years of daily prices between 1998 and 2013 and was validated on the out-of-sample period between 2014 and 2017. The major hypothesis verified in this paper is that machine learning methods select strategies with evaluation criterion near the highest one, but in significantly lower execution time than the brute force method (Exhaustive Search).
Źródło:
Central European Economic Journal; 2018, 5, 52; 206 - 229
2543-6821
Pojawia się w:
Central European Economic Journal
Dostawca treści:
Biblioteka Nauki
Artykuł
Wyświetlanie 1-10 z 20

Ta witryna wykorzystuje pliki cookies do przechowywania informacji na Twoim komputerze. Pliki cookies stosujemy w celu świadczenia usług na najwyższym poziomie, w tym w sposób dostosowany do indywidualnych potrzeb. Korzystanie z witryny bez zmiany ustawień dotyczących cookies oznacza, że będą one zamieszczane w Twoim komputerze. W każdym momencie możesz dokonać zmiany ustawień dotyczących cookies